There are needs for both high resolution imaging and high sensitivity detection/analysis of surface chemistry on a
nanometer scale. These needs can be addressed with Raman spectroscopy coupled with schemes that provide
extraordinary enhancement of the Raman signal, namely surface enhanced (SERS) and tip enhanced Raman
spectroscopy (TERS). Advances in applications of high resolution imaging and high sensitivity detection will be
enabled by two specific improvements: increased signal enhancement and increased robustness of the plasmonic
structures needed to achieve enhancement. Robustness and stability are especially important for those plasmonic
structures made of silver that usually provide the best enhancements. Here we focus particularly on TERS, in which a
plasmonic structure is placed on a scanning probe microscope tip in order to achieve high lateral resolution imaging. We
have demonstrated that aluminum oxide protected silver plasmonic structures show significantly increased robustness
against chemical and mechanical degradation when compared to unprotected analogues without loss of enhancement. A
2-3 nm thick coating of aluminum oxide prevents chemical attack of the underlying silver film for three months in a
desiccator, significantly increasing the storage life of current probes. The same protective coating also extends the
scanning life of the probe when the probe is used to image a hard patterned silicon substrate.
High resolution chemical imaging of surfaces can be achieved using Tip Enhanced Raman
Spectroscopy(TERS), an emerging technique that combines scanning probe microscopy with optical spectroscopy and
takes advantage of apertureless near-field optics to obtain lateral resolution dramatically better than that provided by
conventional optics. So far a 20 nm lateral resolution in chemical imaging of a surface has been achieved. The
plasmonic structures on the tip used for imaging could also be used for novel, high sensitivity, local chemical and
biological sensing. However, the silver plasmonic structures suffer from limited lifetimes due to morphological changes
resulting from heating, wear during imaging, and tarnishing.
The lifetimes of silver plasmonic structures on flat surfaces (as model systems) and on silicon nitride TERS tips
have been extended by depositing over the silver an ultrathin (3nm) silicon oxide (SiOx) coating. With this thickness
protective coating, the contrast factor for the tip, which is the key parameter controlling one's ability to image with the
tip, is decreased slightly (~10%) initially, but the rate at which the signal enhancement degrades is sharply reduced. The
silver layer on an unprotected tip was mechanically damaged after only three images of a polymer surface, while a silver
layer protected by SiOx remained intact after scanning three images.
Several technologies have attempted to deliver the analytical capabilities of Raman and fluorescence spectroscopies to developing nanotechnologies. They have, however, two limitations when applied to nanoscale structures: (i) diffraction limit and (ii) weak signal due to a small sampling volume. To overcome the first obstacle, researchers traditionally use aperture-limited near-field optics based on optical fibers with extremely small apertures (down to ~50 nm). Low transmission through the apertures exacerbates the second limitation by strongly decreasing the measured optical signal. An alternative method based on plasmon optics, strong and very local enhancement of the electric field of light in the vicinity of plasmon nanoparticles (usually Ag or Au), helps to overcome both problems. We overview developments in apertureless near-field optics that are based on a combination of optical spectroscopy and scanning probe microscopy (SPM), with SPM tips modified to have plasmon resonance at the apex. Apertureless near-field microscopy enables traditional confocal optical imaging, scanning probe microscopy (SPM), and a combination of optical and SPM imaging with spatial resolution ~10-20nm, unprecedented for optical techniques. We demonstrate simultaneous Raman and SPM imaging of semiconductor structures and also discuss the challenges facing widespread applicability of this emerging technology, for areas as far ranging as biomedical, semiconductor, and composite materials research.
Tip-enhanced Raman spectroscopy (TERS) using side illumination is a promising spectroscopic tool for nanoscale characterization of chemical composition, structure, stresses and conformational states of non-transparent samples. Recent progress has shown signal enhancements for a variety of samples, including break-through enhancements of semiconductors. In this work, optimization of the polarization geometry increases contrast between near-field and far-field signals on Si and improves imaging quality. Two-dimensional images of semiconductor nanostructures show reasonable agreement between topographical and TERS images. These recent TERS results using both silver- and gold-coated tips demonstrate localization of the Raman enhancement to within approximately 20 nm of the tip. Also, the enhanced Raman signal of a strained Si layer is separated from an underlying Si substrate, which is encouraging for potential strain distribution analysis of silicon nanostructures.
In contrast with aperture-limited Scanning Near-field Optical Microscopy, where the focusing of light is achieved only with very high attenuation, in apertureless near-field optics light is both focused and strongly amplified by the surface plasmons of the probe. Although the general feasibility of this idea and the unprecedented in optics lateral resolution of ~ 15-30 nm have already been demonstrated, the actual field enhancement has so far been well below theoretical expectations, and the useful optical signals have been weak. To bridge the gap between the "proof-of-concept" experiments and reliable optical microscopy with molecular-scale resolution, one needs to unify accurate simulation with effective measurements of the optical properties of the tips and with fabrication. We use dark-field microscopy with side collecting optics for measurements of the optical properties of the tip. The side view allows us to observe the radiation of the tip and hence to analyze its optical properties at the apex. In addition, the measured Raman signal provides an estimate of the electric field enhancement by the tip. Our simulation protocol consists of two parts: electrostatics and wave analysis. Electrostatic simulations give good qualitative predictions, are very fast and therefore conducive to multiparametric optimization. Full wave analysis is needed to evaluate the dephasing effects and far-field signals. The Finite Element Method is used for
all simulations. Various tip designs with the field enhancement ranging from ~ 50 to over 250 (depending on various parameters),
with the commensurate enhancement of the Raman signal by ~ 454 (for gold coating) and ~ 2704 (for silver coating), are presented and analyzed.
Tip-enhanced Raman spectroscopy (TERS) is emerging as a promising spectroscopic tool for nanoscale characterization of chemical composition, structure, stresses and conformational states. However, its widespread application requires optimization of the technique to reproducibly achieve sufficiently high contrast between near-field and far-field signals. We present a TERS spectrometer, based on side illumination geometry, which demonstrates reproducible enhancements of the Raman signal of the order of 103-104 for a variety of molecular, polymeric and semi-conducting samples using both silver- and gold-coated tips. We estimate the localization of the Raman signal enhancement to be ~20 nm. For thick samples, the contrast is limited by a strong far-field signal (from the laser illuminated spot) that overpowers the near-field signal (enhanced in the vicinity of the tip). Optimizing the polarization geometry and the incident angle, we have achieved a contrast between near-field and far-field signal of 12 times on (100) Si - a level that makes this technique attractive for characterization of silicon nanostructures.
The local electric field enhancement in the vicinity of a metal-coated or metal tip is a significant factor in the performance of apertureless near-field optical microscopy and spectroscopy techniques. Enhancement, which is related to the generation of localized surface plasmons in the metal tip, can be maximized when the plasmons resonate at the probing wavelength. Thus the resonance frequencies of the tip apex are crucial to near-field optics. However, it remains a challenge to measure the optical properties of the apex of a tip with a radius much smaller than the wavelength of light. A dark-field scattering spectroscopy method is presented in combination with a side-illumination nano-Raman spectrometer to experimentally determine the optical properties of the tip. The dependence of the optical resonance on the metal deposited is shown for silver- and gold-coated tungsten tips as well as gold-coated silicon nitride tips. The enhancement for Si using gold-coated silicon nitride tips is somewhat larger for a wavelength of 647 nm than for a wavelength of 514.5 nm. The former is closer to the plasmon resonance observed for this tip at ~680 nm.
Several classes of computational methods are available for computer simulation of electromagnetic wave propagation and scattering at optical frequencies: Discrete Dipole Approximation, the T-matrix - Extended Boundary Condition methods, the Multiple Multipole Method, Finite Difference (FD) and Finite Element (FE) methods in the time and frequency domain, and others. The paper briefly reviews the relative advantages and disadvantages of these simulation tools and contributes to the development of FD methods. One powerful tool - FE analysis - is applied to optimization of plasmon-enhanced AFM tips in apertureless near-field optical microscopy. Another tool is a new FD calculus of "Flexible Local Approximation MEthods" (FLAME). In this calculus, any desirable local approximations (e.g. scalar and vector spherical harmonics, Bessel functions, plane waves, etc.) are seamlessly incorporated into FD schemes. The notorious 'staircase' effect for slanted and curved boundaries on a Cartesian grid is in many cases eliminated - not because the boundary is approximated geometrically on a fine grid but because the solution is approximated algebraically by suitable basis functions. Illustrative examples include problems with plasmon nanoparticles and a photonic crystal with a waveguide bend; FLAME achieves orders of magnitude higher accuracy than the standard FD methods, and even than FEM.